Three decades of anaplastic lymphoma kinase: from physiological roles to cancer and immune evasion

Abstract

Anaplastic lymphoma kinase (ALK) is a receptor tyrosine kinase with structural criteria shared among members of the insulin receptor subfamily. ALK plays crucial physiological roles such as in neural tissue development as well as in thinness and body weight regulation. In addition, ALK is a key driver of systemic inflammation that may lead to lethal sepsis. The oncogenic potential of ALK was first described in 1994 when 2 independent research groups discovered that ALK is aberrantly expressed in a rare aggressive subtype of T-cell lymphoma descriptively known as anaplastic large-cell lymphoma. Later, it was found that the aberrant expression of ALK extends to include other neoplasms such as non-small cell lung cancer, neuroblastoma, inflammatory myofibroblastic tumors, and others. In these neoplasms, ALK plays a central oncogenic role by upregulating the activity of a comprehensive network of upstream and downstream survival systems, which causes tumor cell survival and proliferation. The development of selective ALK inhibitors has led to paradigm shift in how ALK-expressing tumors, particularly non-small cell lung cancer, are treated. More recently, ALK has been implicated in disturbing the integrity of the immune system and sustaining immune evasion. In this review, we highlight 3 decades of ALK research, briefly discuss its physiological and pathological roles with a focus on cancer and immunity, and provide an update on the clinical utilization and mechanisms of resistance to ALK inhibitors.

ALK structure

Anaplastic lymphoma kinase (ALK) is a versatile receptor tyrosine kinase with important physiological functions, primarily related to neural tissue development and thinness; it is also the central instigator of survival in several types of cancer [1–4]. The human ALK gene, located on chromosome 2p23, encodes for a polypeptide with 1620 amino acids that undergoes post-translational modifications to produce the mature ALK protein [5, 6] (Fig. 1). ALK has an external ligand-binding domain, a transmembrane domain, and an intracellular tyrosine kinase domain [5–8]. The kinase domain of ALK spans Leu1116 to Thr1392, and contains 3 critical tyrosine residues at positions Tyr1278, Tyr1282, and Tyr1283, which form part of the YXXXYY motif within the activation loop [9, 10]. This motif acts as the major autophosphorylation site of ALK, and is a common structural component among a relatively small family of receptor tyrosine kinases that also includes the insulin receptor, type I insulin-like growth factor receptor (IGF-IR), and leukocyte tyrosine kinase (LTK). Similar to other receptor tyrosine kinases, ALK has docking sites where it recruits downstream proteins such as IRS-1 (Tyr1096), Shc (Tyr1507), and PLC-γ (Tyr1606), which upon their phosphorylation act as “joining” sites to propagate ALK signaling to downstream tyrosine and serine/threonine kinases and transcription factors [11–13]. Nonetheless, IRS-1, Sch, and PLC-γ are only examples because ALK has a complex network of downstream molecules that interact to propagate its signaling [14, 15]. More detail on ALK’s signaling network is discussed below.

Fig. 1.

ALK protein structure and ligand-induced activation. ALK, a receptor tyrosine kinase, has an extracellular ligand-binding domain, a transmembrane (TM) domain, and an intracellular protein tyrosine kinase (PTK) domain. The extracellular domain (1038 residues) includes the following 6 regions: (1) 2 copies of meprin–A-5 protein–receptor protein tyrosine phosphatase μ (MAM); (2) low-density lipoprotein receptor class A (LDLα); (3) glycine-rich or G-rich loop; (4) tumor necrosis factor-alpha (TNFα)-like region; and (5) epidermal growth factor (EGF)-like region. The TNFα-like and G-rich regions intersect several times. The intracellular domain of ALK is its protein tyrosine kinase, which structurally consists of a G-rich loop, α-C helix, catalytic loop, and activation loop (A loop). Several crucial amino acid residues are located within the intracellular domain. Upon ligand, ALKAL1 and ALKAL2, binding, ALK dimerization and subsequent phosphorylation occur

Essential to ALK function and tyrosine kinase activity is the ATP binding pocket located within the kinase domain. The ATP binding pocket contains several amino acid residues that possess specified biological functions, such as His1124 (stabilizes ATP molecule and regulates kinase activity), Lys1150 (orients the phosphate groups of the ATP molecule), Glu1167 (interacts with Lys1150 to lock the kinase domain into active orientation), Ile1171 (influences the shape of the ATP pocket), Leu1196 (gatekeeper that regulates access to the catalytic site), Glu1197 (critical hydrogen bond acceptor; binds ALK inhibitors), Gly1202 (binds ALK inhibitors), His1247-Arg1248-Asp1249 (HRD motif; facilitates transfer of phosphate groups from ATP molecule to tyrosine residues on protein substrates, and Asp1270-Phe1271-Gly1272 (DFG motif; coordinates Mg²⁺ ion binding required for ATP orientation) [7, 16–18].

Mechanisms of ALK activation

The first step in ALK activation is the binding of its extracellular domain with its ligands, which causes dimerization and subsequent phosphorylation of the kinase domain (Fig. 1). Although earlier studies identified heparin and the heparin-binding growth factors, including pleiotrophin and midkine, as ALK ligands, these ligands are nonspecific for ALK activation, and ALK has historically been considered an “orphan receptor” [19, 20]. More recent studies found that ALKAL1 (Augmentor β, FAM150A) and ALKAL2 (Augmentor α, FAM150B) specifically bind and activate ALK [3, 21, 22]. The cytokine-binding sites of ALK have a novel architecture that interacts with ALKAL1 and ALKAL2. In this regard, a new model for ALK activation that entails receptor tyrosine kinase dimerization through monomeric ALKAL1 or dimeric ALKAL2 ligands was proposed. It was shown in this model that following ligand binding, the extracellular domain of ALK undergoes significant rearrangement, aligning parallel to the cell membrane. Interaction between the ligand and the membrane reinforces this orientation [23].

Physiological roles of ALK

ALK is crucial for the growth and function of the nervous system, where it regulates the fundamental processes of cell division, survival, and differentiation in response to extracellular stimuli. In the early stages of human development, ALK is found in all parts of the central nervous system including the ganglia, olfactory bulb, midbrain, and thalamus [1–3, 5]. Studies in mice have shown that ALK deficiency causes a decrease in neurogenesis. The expression of ALK and LTK in the hippocampus and amygdala is linked to memory and anxiety. Intriguingly, mice lacking ALK show reduced anxiety, whereas LTK-deficient mice do not display significant cognitive alterations [24]. Moreover, ALK expression appears to regulate alcohol consumption and behavioral response to cocaine [25, 26]. Recent studies have shown that the interaction of ALK and its ligand ALKAL2 results in pain signaling modulation [27]. ALK has also been found in various ocular tissues, but its exact role is not yet known [28].

ALK have been shown to regulate thinness in adults and children. A recent genome-wide association study conducted in a healthy Estonian population identified an association between thinness and genetic variants in the first intron of the ALK gene [29]. To confirm this association, they evaluated different genetic mouse models and demonstrated that ALK knockout leads to a thin phenotype that is resistant to diet-induced obesity [29]. Furthermore, in another study, knockout of the ALK ligands ALKAL1 and ALKAL2 was associated with the thinness phenotype and adipose tissue mass reduction [30].

ALK mRNA has also been found in the ovaries and testes, which suggests that ALK may also have physiological roles in the development and/or homeostasis of the reproductive system [28]. Knockout of ALK has been associated with delayed puberty and breeding in mice [31]. Moreover, the ALK inhibitors crizotinib and ceritinib have been linked with endocrine side effects including reduced testosterone level in young human males [32].

Aberrant expression and oncogenic potential of ALK in cancer cells

The importance of ALK in carcinogenesis has been widely studied since it was discovered to be a partner in the chimeric oncogenic protein NPM-ALK in T-cell anaplastic large-cell lymphoma (ALCL; NPM-ALK + T-cell lymphoma) [33]. Subsequently, a number of other malignancies have been found to have aberrant ALK expression via translocation, mutation, amplification, or alternative splicing.

Cellular transformation driven by constitutive activation of ALK can result from its oligomerization with different fusion partners (Fig. 2). ALK’s carcinogenic activity is also associated with the presence of point mutations that activate the kinase. T-cell lymphoma, non-small cell lung cancer (NSCLC), neuroblastoma, renal cell carcinoma (RCC), inflammatory myofibroblastic tumor (IMT), rhabdomyosarcoma, and breast cancer are only some of the cancers that harbor these genetic changes.

Fig. 2.

Mechanisms of ALK activation. There are 2 major mechanisms that lead to the activation of ALK: (1) physiological activation of wildtype ALK via ligand-dependent stimulation or (2) aberrant activation of ALK through formation of ALK fusion proteins, isoforms, amplification or mutations in the protein tyrosine kinase region. Notably, the ALK fusion proteins and ALK isoforms contain the protein tyrosine kinase region and are localized intracellularly. ATG 1069, ATG 1071, and ATG 1089 are examples of translational start codons that induce the expression of ALK isoforms

In addition, ALK splicing isoforms have been discovered in NSCLC [34]. However, their mechanistic role in tumor development is still unclear. One study reported a carcinogenic ALK isoform, due to an alternative transcription initiation site, in 11% of melanomas and other tumor types [35]. Studies conducted in vitro and in vivo demonstrated that this isoform occurs independently from other ALK genomic changes, and the ALK isoform-driven cancers are responsive to ALK inhibitors [35].

Non-transcriptional mechanisms of upregulation of ALK expression

Whereas mechanisms of transcriptional upregulation of ALK expression in cancer cells—ALK gene translocations, amplifications or mutations—have been documented in numerous studies, posttranscriptional mechanisms that lead to the upregulation of ALK protein expression have been less frequently investigated.

MicroRNA

MicroRNA (miR) are small non-coding RNA that control a variety of vital cellular processes. They bind to the 3′-untranslated region (UTR) of target mRNAs to block protein translation, which results in posttranscriptional repression of protein synthesis. In studies in our lab, we found that miR-96 binds to ALK, which leads to decreased expression of NPM-ALK in T-cell lymphoma, EML4-ALK in lung cancer, and full-length ALK in neuroblastoma, which suggests that miR-96 is a universal suppressor of ALK expression [36]. In support of this idea, the expression of miR-96 was remarkably reduced in these neoplasms compared to its levels in their respective normal cellular counterparts [36].

In addition to miR-96, other miRs have been implicated in the regulation of ALK expression. For example, miR-424-5p and miR-503-5p have been shown to regulate ALK expression in neuroblastoma. Whereas only miR-424-5p exhibited direct binding to ALK 3′-UTR, the 2 miR induced significant reduction in ALK protein levels as well as inhibition of cell viability in ALK + neuroblastoma cell lines [37].

Heat shock proteins

The molecular chaperones heat shock protein-70 (Hsp70) and Hsp90 are also necessary for the NPM-ALK oncoprotein’s stability and function. [38, 39]. Reduced NPM-ALK expression and association with Hsp90 was seen after NPM-ALK + T-cell lymphoma cells were treated with the Hsp90 inhibitor 17-allylamino-17-demethoxygeldanamycin (17-AAG), likely because the inhibition of Hsp90 made NPM-ALK a target for proteasomal degradation through increased association with Hsp70 [38]. In addition, 17-AAG decreased the association between NPM-ALK and Shc, Grb2, and IRS-1. Moreover, it has also been demonstrated that the activity of matrix metalloproteinase (MMP)−9, an enzyme that breaks down extracellular matrix to facilitate tumor cell invasion, in NPM-ALK + cells depends on Hsp90 [40].

EML4-ALK was also found to be associated with Hsp90. The Hsp90 inhibitor IPI-504 causes rapid degradation of EML4-ALK and subsequent inhibition of downstream signaling pathways, which results in tumor suppression. The expression of the fusion protein sensitizes the cells to IPI-504 both in vitro and in vivo as demonstrated by the transfection of EML4-ALK cDNA into HEK-293 cells [41].

SUMOylation

SUMOylation is a posttranslational modification that is characterized by the reversible conjugation of small ubiquitin-like modifiers (SUMO) and target proteins. SUMO competes with ubiquitin for protein binding, and SUMOylation thus protects target proteins from proteasomal degradation. We previously showed that SUMOylation is deregulated in NPM-ALK + T-cell lymphoma cell lines and primary tumors [42]. We also identified Lys24 and Lys32 within the NPM domain as the sites where NPM-ALK binds with SUMO-1 and SUMO-3. Notably, inhibiting SUMOylation using sentrin/small ubiquitin-like modifier-specific protease 1 (SENP1) decreased the accumulation of NPM-ALK and induced significant negative effects in these lymphoma cells. Treating the cells with SENP1 was also associated with the increased association of NPM-ALK with ubiquitin, which facilitated its degradation [42].

ALK signaling

Physiologically, ALK is a dependence receptor, i.e., it is activated in the presence of its ligands. However, in the absence of the ligands, ALK is degraded by caspase 3 during cell apoptosis [43]. ALK activation can promote several important cell survival-sustaining activities including cell proliferation, transformation, and anti-apoptotic signaling. Four major pathways that ALK and its fusion proteins activate and interact with are (1) Janus kinase (JAK)-signal transducer and activator of transcription (STAT); (2) phosphoinositide 3-kinase (PI3K)-AKT (protein kinase B); (3) RAS-mitogen-activated protein kinase (MAPK); and (4) phospholipase C-gamma (PLC-γ) (Fig. 3). These pathways are extensively characterized and known to regulate oncogenic processes such as apoptosis resistance, cell cycle progression, proliferation, and therapeutic resistance [44]. In addition to these pathways, the aberrant expression of full-length ALK and its fusion proteins activate and interact with other signaling pathways (mentioned below) that play central roles in survival and drug resistance in ALK + neoplasms.

Fig. 3.

ALK survival signaling. The phosphorylation and subsequent activation of ALK triggers several downstream signaling pathways that collectively induce cell survival and proliferation. ALK, similar to other receptor tyrosine kinases, can also bind and interreact with other oncogenic receptors, e.g., IGF-IR and TrkA, when aberrantly expressed. This interaction enhances ALK constitutive phosphorylation and activation and augments cell survival and drug resistance

JAK-STAT

NPM-ALK + T-cell lymphoma has been frequently used in experimental, preclinical, and clinical research to dissect the mechanisms underlying STAT3 oncogenic signaling, and to identify and characterize its downstream partners owing to the high levels of expression of STAT3 and its activated form pSTAT3 [45]. Indeed, STAT3 is considered an eligible therapeutic target in these lymphoma cells [46]. In one model for STAT3 activation, NPM-ALK directly phosphorylates STAT3 at Tyr705 [47, 48]. In another model, STAT3 can also be activated via JAK3, which is constitutively activated by NPM-ALK [47, 48]. JAK3 can also be activated via interaction with the IL-2 family of cytokines such as IL-9 and IL-21, the physiological activators of JAK3 [49, 50]. In support of this idea, pJAK3 expression is upregulated in NPM-ALK + T-cell lymphoma and its selective inhibition causes apoptosis and cell cycle arrest [51–53]. A recent study provided new insights into the mechanisms by which STAT3 drives the survival machinery in NPM-ALK + T-cell lymphoma [54]. Activated STAT3 interacts with key transcription factors, named the transcriptional core regulatory circuit (CRC), which include transcription factors BATF3, IRF-4, and IKZF1. The interactions between STAT3 and the CRC appear to govern the gene expression program in this lymphoma.

Phosphorylated STAT3 homodimerizes and then translocates to the nucleus, where it orchestrates the transcription of genes encoding proteins that are key regulators of apoptosis, the cell cycle, and cell motility, with a net result of cell survival and cancer cell dissemination. Such proteins include cyclin D1, MCL-1, BCL-2, BCL-X_L, survivin, and c-MYC [55]. Recent studies identified additional downstream molecules that are regulated by STAT3 signaling in NPM-ALK + T-cell lymphoma, such as HES1, TXNIP, and SBNO2 [56, 57]. These molecules contribute significantly to the cell survival and anti-apoptotic effects of STAT3 in this lymphoma. Moreover, in NPM-ALK + T-cell lymphoma, STAT3 is responsible for suppressing the expression of T-cell identity genes like CD3e, ZAP-70, LAT, and SLP-76 [58].

STAT3 also promotes tumor cell survival via regulation of miR expression. For instance, STAT3 causes the activation of the miR cluster 17 ~ 92, which was found to promote the proliferation and survival of NPM-ALK + T-cell lymphoma cells [59]. In contrast, STAT3 induces upregulation of DNMT1, a DNA methyltransferase that epigenetically silences several genes including the tumor-suppressor miR-150 [60]. The repression of miR-150 is responsible for ALK + tumors resistance to crizotinib, which is mediated through upregulating the expression of MYB, a target of miR-150. Hence, a mechanism of ALK-STAT3-mediated tumor survival occurs via the repression of miR-150 [60]. Studies from our lab also showed that the NPM-ALK-STAT3 axis suppresses the expression of the MIR26A1 gene that encodes miR-26 [61]. In turn, miR-26 induces posttranscriptional downregulation of iNOS, the enzyme that generates the highly active free radical nitric oxide in NPM-ALK + T-cell lymphoma [61].

Moreover, in NSCLC, the expression of EML4-ALK results in the activation of the JAK2, STAT1, STAT3, STAT5, and STAT6 signaling pathways [62]. The mechanisms driving STAT3 activity and subsequent transcriptional regulation of apoptosis in residual tumors of ALK-rearranged lung cancers are key drivers of treatment-induced adaptive cancer cell survival [63].

PI3K-AKT

By binding to the PI3K regulatory subunit (p85), NPM-ALK initiates the PI3K signaling cascade in NPM-ALK + T-cell lymphoma through activation of the p110 catalytic subunit [64, 65]. Activated PI3K phosphorylates phosphatidylinositol 4,5-bisphosphate (PIP2), which leads to the production of the lipid second messenger PIP3 [66]. Thereafter, PIP3 recruits the serine/threonine protein kinase AKT to the plasma membrane where it undergoes phosphorylation at Thr308 by serine/threonine protein kinase PDK1 [67, 68]. For the maximum activation of AKT, the phosphorylation of Ser473 subsequently occurs via mTORC2 [68]. In NPM-ALK + T-cell lymphoma, PI3K/AKT/mTOR signaling pathway sustains cell growth and survival. Notably, STAT3 and JUNB/CJUN transcription factors, which are highly expressed in NPM-ALK + T-cell lymphoma, are considered to be major regulators of the expression of PI3K, AKT, and mTOR genes [69].

Constitutive activation of the PI3K-AKT-mTOR signaling by NPM-ALK promotes cell survival through the suppression of pro-apoptotic pathways, inhibition of cell cycle progression, and enhancement of protein synthesis and cellular growth [70, 71]. When NPM-ALK + T-cell lymphoma cells were treated with PI3K-AKT-mTOR signaling inhibitors, apoptosis was triggered and proliferation was decreased, in addition to cell cycle arrest [72, 73]. Mechanistically, this pathway causes cell cycle progression and inhibits apoptosis in NPM-ALK + T-cell lymphoma via downregulation of the cyclin-dependent kinase inhibitor p27 and the phosphorylation and nuclear exclusion of the FOXO3a (FKHRL1) tumor suppressor transcription factor [73–75].

Furthermore, PI3K-AKT signaling plays important roles in promoting the survival of other forms of cancer that harbor the ALK oncogene. In EML4-ALK + NSCLC cells, it was found that coadministration of PI3K inhibitors with ALK inhibitors enhances the inhibitory response of the latter [76].

In neuroblastoma with the ALK^F1174 mutation, ALK inhibitors like crizotinib reduced the activity of PI3K as well, further confirming PI3K as one of the downstream signaling pathways of ALK as well as its relevance as a survival-promoting molecule in the different types of ALK + tumors [77]. In addition to activation by ALK, it was recently shown that PI3K-AKT-mTOR pathway can also be activated by IGF-IR signaling in neuroblastoma [78].

RAS-MAPK

It was previously demonstrated that the NPM-ALK oncogenic protein constitutively activates the RAS-MAPK-ERK pathway [79, 80], which in turn upregulates cyclin D1 and promotes CDK4 activity leading to phosphorylation and inactivation of the retinoblastoma protein. In addition, the JUNB transcription factor and the serine/threonine kinase mTOR are two crucial downstream mediators of MEK/ERK signaling in NPM-ALK + T-cell lymphoma [81]. Collectively, this cascade preserves survival by favoring the transcription of anti-apoptotic genes. Thus, activated RAS-MAPK-ERK pathway is crucial for boosting cell proliferation and inhibiting apoptosis in this type of cancer.

In another study in EML4-ALK + NSCLC, the RAS-MAPK signaling pathway was shown to be crucial for the survival of these tumor cells. Simultaneously inhibiting ALK and MEK upfront improved treatment response and prevented resistance in preclinical EML4-ALK + NSCLC models [82].

Previous studies also suggested that RAS proteins collaborate with ALK to transmit signals from the cell surface to the nucleus, and that the ALK-RAS-JNK axis, rather than the classic MAPK pathway, plays a crucial role in neuroblastoma with activated ALK signaling [83].

PLC-γ

PLC-γ, a member of a class of signaling molecules that contains the SH2 domain, is an important ALK downstream signaling pathway. Tyrosine phosphorylation of PLC-γ on Tyr783 and Tyr1254 and subsequent activation of its catalytic domain occur as a result of the complex association between PLC-γ and Tyr664, which is located in the C-terminus of constitutively activated NPM-ALK [84]. The activation of PLC-γ leads to the hydrolysis of PIP2 into the following: (1) the soluble inositol-1,4,5-trisphosphate (IP3), which stimulates the release of Ca²⁺ from the endoplasmic reticulum, and (2) the membrane-bound diacylglycerol, which activates protein kinase C [85, 86].

c-MYC

c-MYC, a proto-oncogene, plays a crucial role in regulating various essential biological processes such as cell cycle progression, cell growth, apoptosis, metabolism, and biosynthesis. c-MYC has been shown to be upregulated by STAT3 in several cancer cell types. Studies have found c-MYC downregulation, with STAT3 knockout, in NPM-ALK + T-cell lymphoma, which suggests that c-MYC is a downstream target of NPM-ALK through the STAT3 signaling pathway [55, 87]. Using a loss-of-function, whole-genome, short-hairpin (shRNA) screen, researchers identified c-MYC binding protein, MYCBP, as a factor affecting the response to crizotinib treatment in EML4-ALK + NSCLC; this finding demonstrates that ALK influences the transcriptional expression of and activates c-MYC, which in turn regulates target genes [88]. A more recent study showed that in cancer stem-like cells of NPM-ALK + T-cell lymphoma, NPM-ALK heterodimers in the nucleus block proteasomal degradation of MYC and increase its accumulation [89]. In another study, coactivation of ALK and N-MYC (ALK F1174C/N-MYC) in a mouse model was shown to cause neuroendocrine prostate cancer via the Wnt/β-catenin signaling pathway [90].

IGF-IR

Like ALK, IGF-IR is a member of the insulin receptor subfamily of receptor tyrosine kinases. IGF-IR is highly expressed in different types of cancer, where it has been documented to promote cancer cell survival and drug resistance. In studies initiated in our lab, we found that IGF-IR is upregulated in NPM-ALK + T-cell lymphoma cell lines and human tumor tissues but not in T lymphocytes and reactive lymphoid tissues [91]. Upregulation of IGF-IR expression in this lymphoma can be attributed, at least in part, to decreased expression of the transcription factors Ikaros isoform 1 (Ik-1) and myeloid zinc finger 1 (MZF1) [92]. Notably, NPM-ALK physically associates with and reciprocally interacts with IGF-IR to sustain their activation/phosphorylation [91, 93]. This reciprocal relationship is supported by the finding that the association between NPM-ALK and IGF-IR was abolished when their lysine residues (Lys210 in NPM-ALK and Lys1003 in IGF-IR, both located in their respective ATP binding domains) were mutated to alanine. We also identified Tyr644 and Tyr664 in NPM-ALK as the residues where IGF-IR binds. The relevance of the associations and interactions between NPM-ALK and IGF-IR was further stressed when their combined inhibition induced synergistic effects in NPM-ALK + T-cell lymphoma [94]. In a similar fashion, combined inhibition of ALK fusion and IGF-IR improved therapeutic efficacy in lung cancer, especially in cases of ALK inhibitor resistance [95].

Insulin receptor substrate-1

Insulin receptor substrate 1 (IRS-1) is a crucial docking protein that acts as the substrate for receptor and cytoplasmic tyrosine kinases [96]. When ALK is expressed in T-cell lymphoma and lung cancer, ALK binds and activates IRS-1 [97, 98]. Subsequently, other proteins bind to phosphorylated IRS-1 to be, in turn, phosphorylated and transmit ALK signaling. Knocking down IRS-1 enhances the anti-tumor effects of ALK inhibitors [95].

Tropomyosin receptor kinase A

The nerve growth factor (NGF) plays important roles in the survival and differentiation of neuronal progenitor cells in the central and peripheral nervous systems [99]. NGF specifically binds and activates the tropomyosin receptor kinase A (TrkA), a receptor tyrosine kinase The NGF/TrkA signaling is essential for the development and maintenance of neural cells particularly the sympathetic and sensory neurons [100, 101]. Whereas the effects of NGF/TrkA signaling in neural tumors vary depending on the tumor subtype, it has invariably survival-promoting effects in non-neural types of cancer. Because of the established role of ALK in neural tissue development, we investigated whether NGF-TrkA signaling plays a role in NPM-ALK + T-cell lymphoma. TrkA was highly expressed in these lymphoma cell lines and human tumors, in contrast to the normal human cellular counterparts including T lymphocytes and reactive lymph nodes [102]. In addition, the lymphoma cell lines secreted significantly higher levels of NGF than the normal T lymphocytes. A selective TrkA inhibitor caused negative cellular effects in vitro, as well as decreased tumor growth, and potentiated the effects of chemotherapeutics in NPM-ALK + lymphoma xenografts. Importantly, we found that TrkA is physically associated with NPM-ALK, and downregulation of ALK using siRNA or an ALK-selective inhibitor caused a substantial decrease in TrkA phosphorylation, and, reciprocally, downregulation of TrkA using siRNA or a TrkA-selective inhibitor led to decreased NPM-ALK phosphorylation. These results suggest that similar to IGF-IR and NPM-ALK, TrkA and NPM-ALK bind and cooperate to promote the survival of NPM-ALK + T-cell lymphoma [102].

Platelet-derived growth factor receptor-beta

In a mouse model of NPM-ALK-triggered lymphoma, the activator protein 1 family members JUN and JUNB have been shown to promote lymphoma development and tumor spread via regulating platelet-derived growth factor receptor-beta (PDGFRB). Inhibiting PDGFRB significantly extended the survival of NPM-ALK transgenic mice and enhanced the effectiveness of an ALK-specific inhibitor in transplanted NPM-ALK tumors [103].

Tissue inhibitors of metalloproteinases

Tissue inhibitors of metalloproteinases (TIMP) are a group of secreted proteins that primarily function to inhibit the degradative activity of MMP. In NPM-ALK + T-cell lymphoma, it has been shown that NPM-ALK-activated STAT3 upregulates the expression of TIMP1 [104]. More recently, TIMP3-ALK fusions have been found in IMT and myxoid fibroblastic tumors of the vocal cord [105, 106], pregnancy-associated IMT of the uterus [107], and rhabdomyosarcoma [108], which implicates the significance of this family in ALK⁺ neoplasms.

Tumors with aberrant ALK expression and activation

Since identification of its oncogenic potential in T-cell lymphoma, the aberrant expression and activation of ALK has been studied in several other types of cancer (Table 1). Studies from different investigators have established a central oncogenic role of ALK signaling in these tumors that entails promotion of cancer cell survival, proliferation, and migration, and inhibition of apoptosis.

Table 1.

Examples of ALK fusion genes in different types of cancer (fusion genes are listed in alphabetical order)

T-cell lymphoma

The oncogenic potential of ALK was first discovered in a rare subtype of T-cell lymphoma (clinicopathologically known as ALCL, which is a misnomer because these lymphoma cells are neither always anaplastic nor always large) [33, 146]. ALK gene fusions occur more frequently in ALCL (60% of cases) than in any other type of cancer. Translocation of the ALK gene (2p23) and the NPM (nucleophosmin; nucleolar phosphoprotein B23) gene (5q35) that induces the chimeric oncogene NPM-ALK is present in 70–80% of all ALK⁺ T-cell lymphoma patients. Numerous NPM-ALK breakpoint variations have been reported [180]. The overexpression of the ALK catalytic domain results from the constitutive expression of the NPM-ALK chimeric protein from the NPM promoter. NPM, as a shuttling protein that translocates between the nucleus and cytoplasm, is responsible for the nuclear localization of NPM-ALK. Research has demonstrated that numerous additional ALK gene rearrangements, including ALO17-ALK, TRK-fused gene (TFG)-ALK, moesin (MSN)-ALK, tropomyosin 3 (TPM3)-ALK, tropomyosin 4 (TPM4)-ALK, 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/inosine monophosphate cyclohydrolase (ATIC)-ALK, myosin 9 (MYH9)-ALK, and clathrin heavy chain (CLTC)-ALK, are also linked to T-cell lymphoma [180]. Several breakpoint variations have been reported in the ALK fusion genes [181–184].

NSCLC

The ALK and EML4 (echinoderm microtubule-associated protein like 4; 2p21) chromosomal translocation was discovered in 2007 in approximately 5% of NSCLC patients [116]. Despite the small percentage of cases, the overall number is large due to the high incidence of NSCLC. The prognosis of patients with ALK rearrangements seemed to improve significantly with the introduction of selective ALK inhibitors to these patients [185].

Neuroblastoma

ALK mutations, specifically F1174C/I/L/S/V, F1245C/I/L/V, and R1275L/Q, have been shown to elicit oncogenic effects in neuroblastoma [186]. Both familial (1–2%) and sporadic (6–10%) neuroblastoma cases harbor ALK mutations, and patients with relapsed disease are more likely to have these mutations [187, 188]. Although several mutations have been described in neuroblastoma, the majority are found in 3 “hot spots” in the ALK kinase domain at residues Phe1174, Phe1245, and Arg1275. These mutations make ALK kinase activation more likely, which results in constitutive activation of downstream survival signaling. In addition to ALK mutations, ALK fusion genes such as TENM3-ALK have also been identified in rare cases of neuroblastoma [189].

IMT

The first solid neoplasm found to have an ALK translocation was IMT. ALK gene fusion with TPM3 or TPM4, 2 genes encoding for a non-muscle tropomyosin, is present in about 50% of IMT cases [190]. Furthermore, more than 10 distinct genes have been shown to be ALK fusion partners in IMT, including CLTC-ALK, ATIC-ALK, SEC31A-ALK, RANBP2-ALK, PPFIBP1-ALK, and CARS-ALK.

Renal cell carcinoma

Renal cell carcinoma (RCC) associated with ALK gene rearrangements is a novel renal cancer entity that accounts for less than 1% of all renal neoplasms [191]. About 40 cases of ALK-rearranged RCC have been documented in patients with ages ranging from 6 to 85 [192]. Although ALK-rearranged RCC exhibits a very diverse range of clinical behaviors, some patients responded dramatically to targeted ALK inhibition; therefore, the identification of this rare entity carries important therapeutic implications [193]. The vinculin (VCL) and ALK gene fusion, which typically occurs in young patients with sickle cell trait [194], the TPM3-ALK fusion, and the EML4-ALK fusion in tumors distinguished by papillary, solid, and mucinous cribriform components, are the most frequent translocations reported in ALK+ RCC patients. Moreover, papillary, solid, tubular, and mucinous cribriform tumors with psammoma bodies frequently include the striatin (STRN) and ALK gene translocation [144].

ALK and the immune system

Recently, growing evidence suggested that ALK plays a role in the homeostasis of the Immune System, which has the potential to affect immunotherapy outcomes in ALK + cancers [195] (Fig. 4). In agreement with this idea, it has been shown that ALK plays important roles in innate, humoral, and cell-mediated immunity as well as in immune evasion.

Fig. 4.

Immune evasion mechanisms in ALK⁺ tumors. ALK-mediated activation of survival-promoting signaling pathways such as JAK/STAT and AKT-PI3K upregulates the expression of PD-L1, which binds PD-1 to induce T-cell exhaustion. Moreover, ALK + tumor cells exhibit decreased expression of human leukocyte antigen, which further facilitates immune evasion. In addition, the activation of ALK oncogenic pathways stimulates tumor-associated macrophages (TAMs) to release IL-10 and TGF-β, which in turn suppress the immune system. Finally, in NPM-ALK + T-cell lymphoma, the expression of CD30 is upregulated and the expression of CD28, CD52, CD69, and TCR are downregulated, with a net result of immune suppression that facilitates tumor establishment and progression

Innate, humoral, and cell-mediated immunity

Stimulator of interferon genes (STING) is essential for effective innate immune response [196, 197]. It induces the expression of IFN-α and IFN-β via upregulation of the NF-κB and IRF3 transcription machinery leading to potent antiviral effects. The ALK pathway is upregulated in patients with sepsis, and has been linked to the rigorous activation of STING pathway. Disruption of ALK expression decreases the STING-mediated immune response to cyclic dinucleotides in monocytes and macrophages, which confers protection against lethal endotoxemia and sepsis in preclinical models. It was also shown that the chimeric protein Mst1/2-ALK promotes NLRP3 inflammasome activation during Listeria monocytogenes infection [198]. In support of ALK role in innate immune response, selective inhibition of ALK reduced sepsis-induced organ insult and death in animal models [199, 200].

Studies of the humoral immune response in pediatric and adult patients with NPM-ALK + T-cell lymphoma reported the presence of circulating anti-ALK autoantibodies in more than 90% of these patients [201, 202]. In previously treated patients, the titers of anti-ALK autoantibodies inversely correlated with significantly lower cumulative incidence of relapse [202]. The anti-ALK autoantibodies have also been detected in other ALK translocation-positive/NPM-ALK-negative T-cell lymphoma patients as well as in other neoplasms with ALK rearrangements such as NSCLC, neuroblastoma, and rhabdomyosarcoma [203–205].

In regard to its potential role in cell-mediated immunity, ALK was found to possess the immunogenic potential to initiate cytotoxic T-cell response in mouse models and humans [206, 207]. Elevated levels of anti-ALK CD8 + T-cells were found in both ALK + T-cell lymphoma patients and healthy individuals [207–209]. Functional analyses of circulating anti-ALK CD8 + T-cells from ALK + T-cell lymphoma patients revealed high levels of both effector and central memory T-cells with pronounced cytolytic activity. In contrast, healthy individuals only had naïve anti-ALK CD8 + T-cells [207]. In mouse models, vaccination with ALK cDNA led to enduring local and systemic protection against NPM-ALK⁺ T-cell lymphoma growth [210]. In addition to CD8 + T-cells, ALK-specific CD4 + T-helper cells have also been detected in ALK + T-cell lymphoma patients, and not in normal subjects, further confirming the cellular immune response against ALK [211]. However, the exact role of ALK in cell-mediated immune response in ALK + malignancies remains incompletely understood and requires further investigation.

Immune evasion mechanisms

Despite evidence that ALK elicits humoral and cellular immune responses, tumors with ALK alterations develop multiple mechanisms to evade the immune system. In the general setting of malignancies, the anti-tumor response of the immune system is impaired. Despite the fact that ALK appears to activate both cellular and humoral immune responses, ALK-mediated immune evasion has been described. For instance, in patients with NPM-ALK + T-cell lymphoma, CD30, a member of the tumor necrosis factor receptor family, is consistently expressed in the lymphoma cells [212]. NPM-ALK has been shown to be associated with the cytoplasmic domain of CD30 and to sustain its expression [213]. In comparison to CD30-negative tumors, CD30 + tumors exhibit downregulation of molecules related to T-cell differentiation and activation, including CD28, CD52, and CD69, as well as TCR signaling [214]. This downregulation can negatively affect the immune response, which impacts the clinicopathological behavior of this lymphoma.

Moreover, NPM-ALK + T-cell lymphoma cells are given a T-regulatory phenotype by activated STAT3, which may help them elude the host’s immune response. In fact, the FOXP3 protein expression, as well as the expression and secretion of IL-10 and TGF-β, are all enhanced by the NPM-ALK fusion protein [215]. These 3 proteins suppress the immune system. Moreover, the immunosuppressive cell-surface receptor programmed cell death ligand 1 (PD-L1; CD274), which binds PD-1 (CD279) and is physiologically expressed on the surface of CD4 + and CD8 + T-cells, is also abundantly expressed by NPM-ALK + lymphoma cells [216]. The interaction between PD-L1 and PD-1 increases the immune response’s ability to induce peripheral T-cell tolerance to self-antigens. The interaction of STAT3 with the PDCDL1 gene promoter significantly increases the transcription of PD-L1 in lymphoma cells, making them able to evade the immune system [216, 217] Similarly, the upregulation of PD-L1 has also been described in EML4-ALK + NSCLC [218, 219]. In NSCLC, EML4-ALK upregulates PD-L1 by activating MEK-ERK and PI3K signaling [219].

Another immune evasion mechanism is ALK’s ability to regulate the antigen presentation machinery, specifically the human leukocyte antigen (HLA) system. For example, in NPM-ALK + T-cell lymphoma mouse models, treatment with ALK inhibitors such as crizotinib and alectinib significantly increased the cell surface expression of HLA-A/B/C [220], which supports ALK-mediated immune evasion via suppression of the cellular expression of the HLA proteins.

Additionally, ALK CAR-T therapies and vaccines are being developed for tumors with ALK mutations [221]. Therefore, a deeper comprehension of the newly discovered functions of ALK in human immune response may offer innovative insights for the creation of the next generation of immunotherapies for malignancies with ALK mutations.

Overall, ALK appears to play a dual role in immune regulation; it can stimulate immune responses and also it can support tumor’s evasion of these responses, stressing how complex ALK’s interaction with the immune system is.

Clinical utilization of ALK inhibitors

The identification of ALK as a central oncogenic protein in different types of cancer has led to a paradigm shift in how these cancers are treated, with ALK-targeted therapies now being included as a standard-of-care treatment. In 2011, crizotinib was the first ALK inhibitor to be approved by the U.S. Food and Drug Administration (FDA) as an upfront treatment for NSCLC [222]. Notably, most of the large clinical studies of crizotinib have been performed in NSCLC patients. Since then, second-generation and third-generation ALK inhibitors have been developed and used in clinical studies. ALK inhibitors abrogate ALK phosphorylation and, subsequently, its tyrosine kinase-based activation of downstream survival proteins. However, the emergence of acquired resistance to ALK inhibitors, which can be caused by a variety of molecular processes like secondary ALK mutations or the activation of other molecular pathways (discussed below), has reduced the effectiveness of these inhibitors [223]. Effective molecular diagnostics are necessary to pinpoint specific resistance mechanisms and direct patients with ALK-rearranged NSCLC to subsequent therapy.

Crizotinib

As the first ALK inhibitor approved for the treatment of ALK-rearranged NSCLC, crizotinib was compared to standard of care first-line treatment pemetrexed plus platinum doublet chemotherapy in a phase III study (PROFILE 1014), and showed efficacy in the first-line setting with superior progression-free survival (PFS) in patients with advanced, previously untreated EML4-ALK + NSCLC [224]. Moreover, crizotinib showed durable responses in patients with ALK + IMT [225], and it was authorized by the FDA in 2022 for use in adults and children with ALK + IMT tumors that are unresectable, recurrent, or resistant to first-line therapy.

Ceritinib

Second-generation ALK inhibitors were developed to enhance the activity against mechanisms of resistance. According to the ASCEND-4 clinical study, ceritinib exhibited strong intracranial activity and a significant PFS benefit that was twice as high as the first-line therapy [226]. The ASCEND-7 study documented the efficacy and safety of ceritinib in EML4-ALK + NSCLC patients with brain or leptomeninges metastasis [227].

Alectinib

In 2015, alectinib, a second-generation ALK inhibitor, received FDA approval to treat EML4-ALK + NSCLC patients whose disease progressed on crizotinib. It was approved in 2017 as a first-line treatment for EML4-ALK + NSCLC by the FDA. In the phase III ALEX trial, alectinib showed a significant benefit in PFS of 34.8 months and a 5-year overall survival (OS) rate of 62.5% when compared to crizotinib, which showed PFS of only 10.9 months and a 5-year OS rate of 45.5%. Moreover, this study showed a lower incidence of adverse effects in the alectinib arm [228].

Brigatinib

Brigatinib is a potent second-generation ALK inhibitor. In the phase III ALTA-1L trial in EML4-ALK + NSCLC patients, brigatinib outperformed crizotinib in terms of survival (3-year PFS: 43% vs. 19%) and efficacy (response rate: 74% vs. 62%). Although ALK inhibitors often have low CNS permeability and are therefore thought to have diminished efficacy in CNS metastasis, a notable finding of this study was that brigatinib-induced PFS improvements were observed in cases with and without CNS metastases [229].

Lorlatinib

Lorlatinib, a third-generation ALK inhibitor, is currently approved as monotherapy for the treatment of adults with advanced EML4-ALK + NSCLC and evidence of disease progression on crizotinib and at least one other ALK inhibitor or on alectinib or ceritinib as first-line therapy for metastatic disease [230]. The FDA and European Medicines Agency approval of lorlatinib for first-line treatment of NSCLC patients was based on the results of the phase III CROWN study, which compared lorlatinib to crizotinib in patients with advanced NSCLC and ALK rearrangement who had not previously received treatment. Disease progression or mortality was 72% less likely in the lorlatinib treatment group than in the crizotinib treatment group [231]. Moreover, lorlatinib had a higher response rate (76%, compared with 58% for crizotinib), as well as a higher intracranial response rate (82%, compared with 23% for crizotinib) [232]. A recent update on 5-year outcome from phase III CROWN study showed that the median PFS was 9.1 months in the crizotinib group, whereas in the lorlatinib group, the median PFS has yet to be reached [233]. Furthermore, the 5-year PFS was 8% and 60% in the crizotinib and lorlatinib groups, respectively [233]. These results indicate that lorlatinib induces the longest PFS reported with any single-agent targeted therapy among all metastatic neoplasms.

Resistance to ALK inhibitors

Resistance to ALK inhibitors falls into 2 main categories: (A) Acquired (secondary to ALK inhibition) and (B) Intrinsic (primary). Notably, most of the studies that addressed resistance to ALK inhibitors were performed in ALK + NSCLC patients.

Acquired (secondary) resistance

This resistance mechanism refers to disease progression after initial clinical response to ALK inhibition. It is further divided into 2 subtypes: ALK-dependent (on-target) and ALK-independent (off-target) (Fig. 5).

Fig. 5.

Mechanisms of resistance to ALK inhibitors. Two major mechanisms of resistance to ALK inhibitors exist: (1) Acquired (secondary) and (2) Intrinsic (primary). Acquired resistance is the most common and includes ALK-dependent (on-target) and ALK-independent (off-target). Intrinsic resistance preexists and develops within 3–6 from start of therapy

1. ALK-dependent (on-target) resistance:

   Secondary ALK activation mutations:

   Thirty percent to 40% of the known resistance mechanisms to ALK inhibition involve the emergence of ALK mutations [223]. These mutations lead to structural changes in the kinase domain that impair the ALK inhibitors’ ability to bind correctly. For instance, treatment with ALK inhibitors results in a larger spectrum of on-target mutations in ALK + NSCLC. The 2 ALK mutations that occur most frequently are L1196M and G1269A, which are found in the ATP deep binding pocket. Previous studies have shown that second-generation ALK inhibitors result in greater curative outcomes and are superior to chemotherapy in NSCLC [234, 235]. However, it was also noted that treatment with second-generation ALK inhibitors is frequently associated with development of resistant ALK mutants. The G1202R mutation, detected in only 2% of patients’ tumor samples after treatment with crizotinib, is the most common cause of acquired resistance to second-generation ALK inhibitors [236]. Indeed, this mutation occurs in 50% of patients receiving brigatinib, and 20%–30% of patients receiving ceritinib or alectinib [237]. Compared with crizotinib, the blood–brain barrier can be crossed by the third-generation ALK inhibitor lorlatinib, which can also block a variety of the ALK resistance mutations. Lorlatinib has proven to have substantial action against common mutations such as L1196M, G1202R, and G1269A [238].

   It was also found that resistance arises when patients simultaneously develop two or more (compound) ALK mutations during treatment with ALK inhibitors, particularly when these inhibitors are used sequentially [239]. Development of the C1156Y/L1198F dual ALK mutant was found to induce resistance to lorlatinib, and in contrast sensitivity to crizotinib [240]. There are several other forms of ALK dual or triple mutants that confer substantial resistance to second-generation ALK inhibitors and third-generation ALK inhibitors [241].

   ALK amplification:

   Although ALK amplification occurs rarely, it has been implicated in the acquired resistance to crizotinib. In one study, 1 of 18 NSCLC patients treated with crizotinib developed ALK amplification [223], and in another study, 2 of 11 patients were found to have ALK copy number gain, with 1 of the resistant patients also developing an ALK inhibitor-resistant mutation [242].

   EML4-ALK fusion variants:

   More than 15 fusion breakpoints in EML4 exons have been identified in EML4-ALK protein [243–245]; EML4-ALK v1 (exon 13 of EML4 fused to exon 20 of ALK, E13;A20) and v3a/b (E6;A20) culminate in 70–80% of the variants, which are followed by EML4-ALK v2 (E20;A20) and v5′ (E18;A20). In 2016, EML4-ALK fusion variant-dependent differential clinical response to crizotinib was first reported in NSCLC patients [246]. Objective response rate (ORR) was 74% in the v1 variant patients compared to 63% in the non-v1 variants group. Of note is that the PFS was 11 and 4.2 months in the v1 and non-v1 patients, respectively. Subsequent studies supported these initial observations and extended the differential response of EML4-ALK v1 vs. non-v1 patients to second-generation ALK inhibitors [247, 248]. Indeed, the crizotinib-resistant ALK G1202R was more frequently found in the EML4-ALK v3 patients than in v1 patients [249]. Hence, it is recommended to treat patients with EML4-ALK v3 with lorlatinib, a third-generation ALK inhibitor, as the first-line therapy option to avoid sequential generation of ALK G1202R and compound ALK mutations [243].

2. ALK-independent (off-target) resistance:

   ALK-independent resistance occurs in 50% of the patients on second-generation ALK inhibitors who develop disease progression. Several off-target mechanisms have been identified.

   Bypass signaling pathways:

   Upregulation of wild-type EGFR activation/phosphorylation was the first bypass signaling pathway identified in the setting of resistance to an ALK inhibitor, crizotinib, in NSCLC [250]. Subsequently, other survival pathways were found to be upregulated and cause resistance to ALK inhibitors. These pathways include KIT/PDGFRA [251], IGF-IR/IRS-1 [252], HER2/3 [253], FAK/YAP1 [254], SRC [255], and BRAF^V600E mutation [256].

   Because crizotinib induces dual inhibition of ALK and MET, the contribution of HGF-MET signaling to off-target resistance was not known until later generations of ALK inhibitors were introduced [257]. GAB1, a target of the HGF/MET signaling, was also shown to lead to alectinib resistance, which was overcome by combining metformin with alectinib [258].

   More recently, using a preclinical experimental approach, a multilevel survival system was shown to initiate apoptotic cell death evasion of drug-tolerant persister cells in NSCLC. This system involves a rapid downregulation of the BH3-only proapoptotic protein NOXA (PMAIP1), a member of the BCL-2 family, leading to persister NSCLC cell dependence on the anti-apoptotic proteins MCL-1 and/or BCL-X_L after treatment with ALK inhibitors. This model also suggests that upregulation of PI3K/AKT and MAPK signaling pathways can hinder the response to lorlatinib [259]. Another recent preclinical study in NSCLC found that the abundant release of GAS6 by tumor-associated macrophages and MMP11-expressing cancer-associated fibroblasts upregulates AXL receptor tyrosine kinase signaling leading to alectinib resistance [260].

   Histopathological transformation:

   This phenomenon is associated with an uncommon form of off-target resistance to ALK inhibitors. It has been observed when NSCLC tumors (adenocarcinoma) transform to small cell or squamous cell carcinoma [261, 262]. The underlying mechanisms for the resistance are not clear. Genetic alterations including mutations of TP53 and PTEN tumor suppressor genes, as well as the expression of the oncogenic proteins BCL-2, CD133, and SOX2 have been implicated in resistance development [263, 264].

   The epithelial-to-mesenchymal transition (EMT) is a dynamic process in which epithelial cancer cells attain morphological features of mesenchymal cells including loss of polarity and intercellular cohesiveness, which makes epithelial cancer cells mobile with enhanced ability to invade surrounding tissues and migrate to form distant metastases. The mechanisms by which EMT induces resistance to ALK inhibition are not completely understood. Although some studies reported EMT in patients who developed resistance to ALK inhibitors, several of these patients were found to simultaneously harbor therapy-resistant ALK mutations [237, 265]. EMT-related signaling molecules that are implicated in the resistance to ALK inhibitors include proteoglycans, HIF-1, FOXO, and ECM receptors [265]. At least one study proposed that the EML4-ALK G1202R mutation in NSCLC initiates EMT-mediated resistance to ceritinib via activation of the STAT3/Slug pathway [266]. Preclinical studies using crizotinib-resistant cells showed that EMT, associated with decreased expression of microRNA-200c and upregulation of ZEB1, caused cross resistance to alectinib, ceritinib, and lorlatinib [267]. Importantly, histone deacetylase antagonism overcame the resistance to ALK inhibitors by reversing EMT. It has been suggested that EMT may evolve subsequently to the resistance to ALK inhibitors. In this regard, a recent study demonstrated that the EML4-ALK L1196M and G1202R mutants induce crizotinib resistance in NSCLC via initiation of an MDM2/MEK/ERK-dependent EMT [268].

   Drug efflux pump:

   P-glycoprotein, encoded by the MDR1 gene, is an ATP-binding cassette multidrug efflux pump that has been implicated in the resistance to the ALK inhibitors crizotinib and ceritinib [269]. P-glycoprotein pumps these ALK inhibitors out of the target cells. In addition, it has been suggested that p-glycoprotein causes poor CNS penetrance of these drugs. Hence, alectinib, which is not a substrate of p-glycoprotein, achieves higher CNS levels in preclinical models of intracranial NSCLC metastases and in patients with EML4-ALK + NSCLC [270, 271].

   Deregulation of metabolic pathways:

   The contribution of deregulated metabolic pathways to off-target resistance to ALK inhibitors has recently gained attention. The ALK fusion proteins NPM-ALK and EML4-ALK upregulate, via a STAT3-dependent mechanism, the expression of the glycolysis regulatory enzyme PFKFB3 in T-cell lymphoma and NSCLC, respectively [272]. In turn, PFKFB3 leads to increased glycolysis in these cancer cells. Importantly, preclinical studies showed that inhibition of PFKFB3 activity overcame the resistance to ALK inhibition [272]. Another recent study showed that the cancer-associated fibroblasts promote resistance to ALK inhibitors by upregulating lipid biosynthesis in NSCLC cells [273]. In vitro experiments showed that dual inhibition of ALK and the lipid-regulatory transcription factor SREBP-1 restored the sensitivity of resistant NSCLC spheroids to ALK inhibition.

Intrinsic (primary) resistance

It is generally agreeable that patients with ALK + tumors who demonstrate disease progression within 3–6 months of the start of ALK inhibitors are considered to have intrinsic drug resistance, i.e., physiological resistance. The exact mechanisms of intrinsic resistance are not systematically studied. In this regard, any of the mechanisms of acquired (secondary) resistance (discussed above), if existed before treatment initiation, are also considered mechanisms of intrinsic resistance. Preexisting molecular aberrancies such as EGFR, KRAS, TP53, and PTEN genes mutations, BIM gene deletion polymorphism, MYC copy number amplification, EML4-ALK rearrangement mutations with a low minor allele frequency, have been among the drivers of the intrinsic resistance to ALK inhibitors [274–276].

One important problem associating crizotinib therapy in NSCLC is intrinsic resistance through intracranial disease progression because of poor penetrance of the blood brain barrier, which results in 384 times higher drug concentration in the peripheral blood than in cerebrospinal fluid [277]. The second- and third-generation ALK inhibitors have been able to overcome this type of resistance.

Conclusion and future perspective

During the past 3 decades, our understanding of how ALK contributes to the survival and progression of different types of cancer has steadily progressed to the extent that today ALK is considered a prototype of oncogenic tyrosine kinases. Numerous studies have demonstrated that ALK is an oncogene that attains a central role in cancer cell survival through the activation of and interaction with a complex network of upstream and downstream tumor survival system. Therefore, ALK is a legitimate target for personalized cancer therapy. Indeed, development of tyrosine kinase inhibitors that selectively target ALK has been the focus of dedicated researchers in academic institutions and pharmaceutical companies. Over relatively few years, several ALK inhibitors have been developed and approved to treat patients who harbor aberrancies of ALK signaling. Nonetheless, despite the existence of several generations of ALK inhibitors, ALK proves to be an elusive molecule, which leads to patients developing resistance via known or yet to be discovered mechanisms.

Future research is expected to lead to improved ALK inhibitors that are more effectively designed. Currently, fourth-generation ALK inhibitors are under development including TPX-0131, a compact macrocyclic molecule designed to fit within the ATP-binding boundary to overcome a variety of ALK mutations in NSCLC [278]. NVL-655 is another macrocyclic fourth-generation ALK inhibitor that binds to the ATP binding pocket of ALK. It showed promising early results that warranted breakthrough designation by the FDA. TPX-0131 and NVL-655 are currently being investigated in the phase I/II clinical trials NCT04849273 (FORGE-1) and NCT05384626 (ALKOVE-1), respectively [279].

Another area for future investigation to improve response to ALK inhibitors is development of more effective strategies of combination therapies that are capable of simultaneously targeting ALK and off-target resistance bypass pathways such as c-KIT/PDGFRA, FAK/YAP1, HGF/MET, and GAS6/AXL. Antibody-drug conjugates, which comprise attachment of a cytotoxic drug and a monoclonal antibody that specifically target ALK via a chemical linker is an example of future strategies to overcome bypass signaling pathway. It is also important to develop strategies to target the ALK-inhibitors persister cells via combining ALK inhibitors with CDK4/6 inhibitors or manipulating the BCL-2 family of proteins. Another approach for future consideration is development of ALK inhibitors that are more effective to treat patients with resistant ALK fusion gene variants, e.g., EML4-ALK v3. Finally, immune system-dependent therapeutic approaches should be developed for future effective treatment of ALK + tumors. For instance, strategies to overcome ALK evasion of the immune response need to be established in order to improve the poor outcome of immunotherapy in patients with ALK + tumors. Furthermore, utilizing ALK as a vaccine target to generate a robust immune response is an attractive therapeutic modality for successful therapy in the future. One way to accelerate the development to immune system-dependent approaches is to expand the patient’s population enrolled in such studies.

An emerging drug technology for future research considerations in ALK + tumors is proteolytic targeting chimers (PROTACs). PROTACs utilize the ubiquitin–proteasome system to degrade target proteins such as ALK. Advantages of PROTACs include low requirements for binding sites, which enhances the accessibility of targeted proteins and lowers the effective doses. Moreover, PROTACs directly degrade target protein without the need to necessarily modify its function [280]. Preclinical studies demonstrated encouraging effects of PROTACs in EML4-ALK + NSCLC [281], but this approach has yet to be tested in clinical trials of patients with ALK + tumors.

Targeting metabolic dependencies and adaptations in ALK + tumors is an intriguing therapeutic modality to explore. Caloric restriction by lowering circulating glucose and growth factor levels through development of structured dieting habits is in line with the idea that metabolic stress of tumors could augment standard treatments such as ALK inhibitors or non-standard treatment such as immunotherapy [282]. Metformin, which lowers blood glucose levels via gluconeogenesis reduction in the liver, has been shown to restore the sensitivity to ALK inhibitors by antagonizing alternate growth-promoting pathways, e.g. IGF-IR and HGF/MET, and inducing energy deprivation lethal stress in ALK + cancer cells [283–285].

Pathological and clinical laboratory testing is an important area to improve the outcome of ALK inhibitors. Repeated tumor biopsies to identify secondary resistance mutations are frequently not feasible as well as carry significant risk to the patients because of the invasive nature of these procedures. Alternatively, ctDNA analysis in plasma is non-invasive and can be used to monitor therapeutic response and drug resistance in patients receiving ALK inhibitors.

The majority of ALK research in cancer, particularly as a therapeutic target, has been focused on NSCLC. Because of the heterogeneity and background variability among different types of cancer, it is crucial to also expand our understanding of ALK biology and its therapeutic potential in other types of ALK + tumors as well. Future preclinical and large-scale clinical investigations that include T-cell lymphoma and other types of solid tumors should help achieve this important goal.

Finally, future research may unravel new physiological roles of ALK, beyond neural tissue development and regulating thinness and body weight. Furthermore, the identification of a major regulatory role of ALK in inflammation and sepsis stresses the possibility of future discoveries of new roles of ALK in pathological conditions other than cancer, and perhaps its potential to become a therapeutic target in such conditions.

Abbreviations

ALCL

Anaplastic large-cell lymphoma

ALK

Anaplastic lymphoma kinase

ALKAL

ALK and LTK ligand

ALO17

ALK lymphoma oligomerization partner on chromosome 17

AML

Acute myeloid leukemia

ATIC

5-Aminoimidazole-4-carboxamide ribonucleotide formyltransferase/inosine monophosphate cyclohydrolase

BATF3

Basic leucine zipper ATF-like transcription factor 3

BCL-2

B-cell lymphoma 2

BCL-X_L

B-cell lymphoma-extra-large

BH3

BCL-2 homology domain 3

BIM

BCL-2 interacting mediator of cell death

BRAF

B-Raf proto-oncogene

CAR-T

Chimeric antigen receptor T-cell

CDK

Cyclin-dependent kinase

CLTC

Clathrin heavy chain

CRC

Core regulatory circuit

LBCL

Large B-cell lymphoma

DNMT

DNA methyltransferase

ECM

Extracellular matrix

EGFR

Epidermal growth factor receptor

EML4

Echinoderm microtubule-associated protein-like 4

EMT

Epithelia-to-mesenchymal transition

ERK

Extracellular signal-regulated kinase

FAM150A/B

Family with sequence similarity 150 member A/B

FKHRL1

Forkhead in rhabdomyosarcoma-like 1

FOXO

Forkhead box O

FOXP3

Forkhead box P3

GAB1

Grb2-associated-binding protein

GAS6

Growth arrest-specific gene 6

GMT

Gastric mesenchymal tumors

Grb2

Growth factor receptor-bound protein 2

HCC

Hepatocellular carcinoma

HIF-1

Hypoxia-inducible factor 1

HLA

Human leukocyte antigen

IGF-IR

Type I Insulin-like growth factor receptor

Ik-1

Ikaros isoform 1

IKZF1

Ikaros family zinc finger 1

IMT

Inflammatory myofibroblastic tumor

IP3

Inositol-1,4,5-trisphosphate

IRF4

Interferon regulatory factor 4

IRS

Insulin receptor substrate

JAK

Janus kinase

JunB

Jun B proto-oncogene

HES1

Hairy and enhancer of split-1

HGF

Hepatocyte growth factor

LTK

Leukocyte tyrosine kinase

MAPK

Mitogen-activated protein kinase

MCL-1

Myeloid cell leukemia-1

MDR1

Multidrug resistance 1 gene

MEK

Mitogen-activated protein kinase

MMP

Matrix metalloproteinases

MSN

Moesin; Mst1/2mammalian Ste20-like protein kinase 1/2

mTOR

Mammalian target of rapamycin

MYCBP

C-MYC binding protein

MYH9

Myosin 9

MZF1

Myeloid zinc finger 1

NLRP3

Nucleotide-binding oligomerization domain-like receptor protein 3

NPM

Nucleophosmin

NSCLC

Non-small cell lung cancer

ORR

Objective response rate

PDAC

Pancreatic ductal adenocarcinoma

PD-1

Programmed death protein 1

PDK-1

3-Phosphoinositide-dependent protein kinase 1

PDK-2

3-Phosphoinositide-dependent protein kinase-2

PD-L1

Programmed death ligand 1

PDGFRA

Platelet-derived growth factor receptor alpha

PFKFB3

6-Phosphofructo-2-kinase/fructose-2,6-biphosphatase 3

PFS

Progression-free survival

PI3K

Phosphoinositide 3-kinase

PIP2

Phosphatidylinositol 4,5-biphosphate

PIP3

Phosphatidylinositol 3,4,5-trisphosphate

PLC-γ

Phospholipase C-gamma

PMAIP1 (NOXA)

Phorbol-12-myristate-13-acetate-induced protein 1

PROTACs

Proteolytic targeting chimeras

RAS

Rat sarcoma

RCC

Renal cell carcinoma

SBNO2

Strawberry notch homolog 2

SCC

Squamous cell carcinoma

SENP1

Sentrin/small ubiqukiritin-like modifier-specific protease 1

SH2

SRC homology 2

SREBP-1

Sterol regulatory element-binding protein 1

STAT

Signal transducer and activator of transcription

STING

Stimulator of interferon genes

STRN

Striatin

TAM

Tumor-associated macrophages

TCR

T-cell receptor

TFG

TRK-fused gene

TENM3

Teneurin transmembrane protein 3

TGF-β

Transforming growth factor-beta

TIMP

Tissue inhibitors of metalloproteinases

TPM

Tropomyosin

TrkA

Tropomyosin receptor kinase A

TXNIP

Thioredoxin-interacting protein

UTR

Untranslated region

VCL

Vinculin

YAP1

Yes-associated protein 1

ZEB1

Zinc finger E-box binding homeobox 1

Author contribution

HMA, SA, MMS, MKK, JH performed literature search and read and approved the contents of the manuscript in its submitted form. HMA, MMS wrote the first draft. HMA, SA wrote the second draft. HMA, SA, MKK, JH prepared the figures. HMA developed the concept, supervised the work, answered the reviewers critique, and performed the edits and revisions of the final draft.

Funding

This work is partially supported by an R01CA260872 grant from the NCI (HMA).

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Competing interests

The authors have no competing financial or non-financial interests to disclose that are relevant to the content of this article.

Clinical trial number

Not applicable.